Video Coding Block Partitioning Reduces Computational Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The enormous number of combinations of CU sizes in the VVC's QTTTBT scheme leads to a significant increase in computation required for evaluating and selecting optimal partition patterns, which is not suitable for coding apparatuses with fixed computation capacity or those requiring reliable processing speed.
Innovation Solution
A coding method that partitions a first block in an original image into a group of second blocks using quad tree, ternary tree, or binary tree partitioning, with two stages of sub-partitioning to reduce the complexity of selecting CU sizes and predict movement for each second block.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the QTTTBT scheme is used to increase the degree of freedom of CU sizes, then coding efficiency is improved, but the number of partition pattern combinations increases enormously leading to excessive computation
Solution Approach 1:
The patent divides the partition pattern selection process into two independent stages: (1) QT partitioning stage that determines the size of parent blocks, and (2) BT/TT partitioning stage that divides parent blocks into second blocks. This segmentation reduces the combinatorial complexity by separating the selection of CU sizes from the selection of partition patterns, making the overall computation manageable while preserving coding efficiency.
Solution Approach 2:
The patent performs preliminary QT partitioning to determine parent block sizes before proceeding to BT/TT partitioning. By pre-determining the sizes of parent blocks based on image characteristics, the subsequent partitioning stage has fewer options to evaluate, thereby reducing the enormous computation that would otherwise be required to evaluate all possible partition patterns.
2Productivity
If all partition patterns are evaluated to select the optimal one, then coding efficiency is maximized, but the processing time increases
Solution Approach 1:
The evaluation process is segmented into two phases: first evaluating QT partition patterns to select parent block sizes, then evaluating BT/TT partition patterns for each parent block. This segmentation allows the system to process partition patterns in a structured manner, reducing the overall processing time compared to evaluating all patterns simultaneously without compromising the selection of optimal patterns.
Solution Approach 2:
The patent performs preliminary evaluation and selection of QT partition patterns before evaluating BT/TT partition patterns. This preliminary action identifies promising parent block configurations, thereby reducing the number of subsequent evaluations needed and decreasing total processing time while still achieving optimal coding efficiency.
3Device complexity
If the CU size selection is adapted to reduce computation, then processing speed becomes variable, but coding efficiency may be compromised
Solution Approach 1:
The patent applies different partitioning strategies to different regions of the image based on local characteristics. By adapting the partitioning approach to local image features while maintaining the two-stage structure, the system reduces computation in suitable regions without compromising overall coding efficiency, as each region receives appropriate processing tailored to its specific characteristics.
Data Source
AI summary
A coding method is a coding method executed by a coding apparatus. The coding method includes partitioning a first block having a predetermined size in an original image into a group of second blocks each being a block serving as a unit for coding, by quad tree, ternary tree, or binary tree, and predicting movement for each of the second blocks. The partitioning includes first sub-partitioning and second sub-partitioning, the first sub-partitioning includes partitioning the first block into blocks each having a size selected from sizes determined according to the quad tree, the ternary tree, or the binary tree, and the second sub-partitioning includes further partitioning the blocks each having the selected size to generate the second blocks.